The panels (a), (b) and (c) in Figure show the VSP, the surface reflection, and the integration tomography result, respectively. The identity operator is used as an integration operator for the integration tomography. Panels (d), (e) and (f) show the ray coverage map for the VSP, the surface reflection and the integration tomography, respectively. The gray scales indicate how many rays pass through the corresponding grids. The surface reflection tomography has a much denser ray coverage than that for VSP tomography (There are totally 12674 rays used for surface reflection tomography and only 4708 rays used for VSP tomography). For the VSP tomography, the rays fall into an approximately upside-down triangle area. Out of this area, the VSP tomography gives a highly deviated inversion result. Since surface reflection tomography has a wider and denser ray coverage, its inversion result is much better than VSP tomography. The sum of ray coverage for the surface reflection and the VSP forms the ray coverage map for the integration tomography. In the areas at surface location smaller than 2.0 km or bigger than 18.0 km, both surface reflection and VSP have poor ray coverage, which results in an unsatisfied integration tomography result in these areas. However, in the area within the anticline, where the VSP tomography does invert velocity more accurately than surface reflection, the integration tomography has a better inversion result than surface reflection tomography.
Figure shows the migration results using the velocity field obtained by VSP, surface reflection, and integration tomography. The left panel corresponds to the VSP tomography; the center panel corresponds to surface reflection tomography, and the right panel corresponds to the integration tomography. Compared to using the surface reflection tomography, notice that using the integration tomography for migration can image the bottom reflectors more accurately, as indicated by two circles in Figure . However, the image using integration tomography is not as accurate as that using surface reflection tomography at the two sides of the model because of the deleterious contribution from VSP tomography in these areas.
Figure shows the three angle domain common image gathers (ADCIG) extracted at same three locations within the anticline when using VSP, reflection, and integration tomography results for migration. The left three panels correspond to VSP tomography; the center three panels correspond to reflection tomography; and the right three panels correspond to integration tomography. From left to right within each group, the corresponding surface locations for three ADCIGs are 8.0, 10.0, and 12.0 km, respectively. We can observe the contributions from both VSP and surface reflection tomography in the ADCIGs for integration tomography.